The aim is to understand the causes of the progressive loss of lean body mass (LBM) which occurs in aging men, with simultaneous increase in adipose mass (AM), centralization of adipose tissue topography, atrophy of muscle mass, loss of muscle strength, decrease of bone mass, and declines in aerobic work capacity and resting metabolic rate. These eight underlined terms are the "dependent" or "outcome" variables in the proposed study. Six potential predictors or risk factors will be investigated: low plasma levels of IGF-I, IGF-II, free testosterone and dehydroepiandrosterone sulfate (DHEAS), high daily excretion of urinary free cortisol, and curtailed physical activity. This latter group of six underlined variables are the "independent" or "predictor" variables. A population of 250 independent overtly healthy men will be monitored for five years by measuring the fourteen variables stated above at six month intervals. The authors will determine by multivariate statistical analyses the role of each independent variable in predicting the age- related decline in each of the dependent variables. Interactions between the dependent variables as predictors of the outcome variables will be identified. Preliminary data of the applicants have already shown a relation of low plasma IGF-I to the loss of LBM, and a relation of low plasma DHEAS to the gain in AM in elderly men. In the proposed study the authors will extend their preliminary findings by including additional hormones capable of influencing body composition, by employing both cross-sectional and longitudinal designs, and by enlarging the study population. The applicants will learn how much of the marked inter- individual variability in the body composition and related physiologic functions of elderly men can be explained by differences in their endocrine status and in their level of physical activity. The clinical significance of the project is as follows. Accelerated loss of LBM with the associated compositional and physiologic changes contribute to frailty and dependency in old age. To the extent that any of the independent variables is found to be a statistically significant predictor of accelerated body composition deterioration in old age, new interventions to prevent or reverse this deterioration will thereby be made available. Thus, it is possible that maintaining a youthful profile of the significant predictor variables in aging men may tend to preserve youthful body composition and related physiologic functions.